[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <51E59FEF.90404@linux.intel.com>
Date: Tue, 16 Jul 2013 12:33:03 -0700
From: Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
To: Greg KH <gregkh@...uxfoundation.org>
CC: Linux Kernel <linux-kernel@...r.kernel.org>,
"Brown, Len" <len.brown@...el.com>,
"Rafael J. Wysocki" <rjw@...k.pl>
Subject: Re: driver model, duplicate names question
On 07/16/2013 12:04 PM, Greg KH wrote:
> A: No.
> Q: Should I include quotations after my reply?
>
> http://daringfireball.net/2007/07/on_top
Glad to learn something new today.
>
> On Tue, Jul 16, 2013 at 11:54:31AM -0700, Srinivas Pandruvada wrote:
>> Hi,
>>
>> I am assigned to do add a powercap class. There are several
>> technologies, which will allow to add a power budget to an individual
>> device. For example, you can set a power budget to a individual
>> physical cpu package, each core and uncore devices, GPUs, DRAM etc.
> "classes" all reference a "device" in the system, I don't see that in
> your tree below, where does that come in? How do I, as someone who
> created a device in the system know to create a your new powercap class
> for it?
>
> In other words, are you _sure_ you want a class here and not something
> else (i.e. a bus?)
>
>> +The Power Capping framework organizes power capping devices under a tree structure.
>> +At the root level, each device is under some "controller", which is the enabler
>> +of technology. For example this can be "RAPL".
>> +Under each controllers, there are multiple power zones, which can be independently
>> +monitored and controlled.
>> +Each power zone can be organized as a tree with parent, children and siblings.
>> +Each power zone defines attributes to enable power monitoring and constraints.
> Ah, this sounds like you want to be a bus, as you have a controller, and
> then devices attached to it.
>
>> +Example Sys-FS Interface
>> +
>> +/sys/class/power_cap/intel-rapl
>> +├── package-0
>> +│ ├── constraint-0
>> +│ │ ├── name
>> +│ │ ├── power_limit_uw
>> +│ │ └── time_window_us
>> +│ ├── constraint-1
>> +│ │ ├── name
>> +│ │ ├── power_limit_uw
>> +│ │ └── time_window_us
>> +│ ├── core
>> +│ │ ├── constraint-0
>> +│ │ │ ├── name
>> +│ │ │ ├── power_limit_uw
>> +│ │ │ └── time_window_us
>> +│ │ ├── energy_uj
>> +│ │ └── max_energy_range_uj
>> +│ ├── dram
>> +│ │ ├── constraint-0
>> +│ │ │ ├── name
>> +│ │ │ ├── power_limit_uw
>> +│ │ │ └── time_window_us
>> +│ │ ├── energy_uj
>> +│ │ └── max_energy_range_uj
>> +│ ├── energy_uj
>> +│ ├── max_energy_range_uj
>> +│ └── max_power_range_uw
>> +├── package-1
>> +│ ├── constraint-0
>> +│ │ ├── name
>> +│ │ ├── power_limit_uw
>> +│ │ └── time_window_us
>> +│ ├── constraint-1
>> +│ │ ├── name
>> +│ │ ├── power_limit_uw
>> +│ │ └── time_window_us
>> +│ ├── core
>> +│ │ ├── constraint-0
>> +│ │ │ ├── name
>> +│ │ │ ├── power_limit_uw
>> +│ │ │ └── time_window_us
>> +│ │ ├── energy_uj
>> +│ │ └── max_energy_range_uj
>> +│ ├── dram
>> +│ │ ├── constraint-0
>> +│ │ │ ├── name
>> +│ │ │ ├── power_limit_uw
>> +│ │ │ └── time_window_us
>> +│ │ ├── energy_uj
>> +│ │ └── max_energy_range_uj
>> +│ ├── energy_uj
>> +│ ├── max_energy_range_uj
>> +│ └── max_power_range_uw
>> +├── power
>> +│ ├── async
>> +│ ├── autosuspend_delay_ms
>> +│ ├── control
>> +│ ├── runtime_active_kids
>> +│ ├── runtime_active_time
>> +│ ├── runtime_enabled
>> +│ ├── runtime_status
>> +│ ├── runtime_suspended_time
>> +│ └── runtime_usage
>> +├── subsystem -> ../../../../class/power_cap
>> +└── uevent
> Ick. Rewrite this to use a bus and you should be fine, right? Don't
> use a class, a class is only to be used if you have a device that is a
> specific "type of thing". Like a tty device, it is a class, as lots of
> different "real" devices can have tty ports on them (usb, pci, pcmcia,
> platform, etc.)
>
> Rethink this using a bus and see if that solves your issues. You get a
> hierarchy with that. And you can have different "types" of devices on
> your bus, making it easy to tell the difference between a "package" and
> a "constraint".
>
> Does that help?
I will experiment your suggestion. I see this class analogous to
"/sys/class/thermal",
, where the thermal class provides a set of consistent interface for all
thermal devices.
> greg k-h
>
Thanks,
Srinivas
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists